The next generation of exoplanet-hunting spectrographs should deliver up to an order of magnitude improvement in radial velocity precision over the standard 1 m/s state of the art. This advance is critical for enabling the detection of Earth-mass planets around Sun-like stars. New calibration techniques such as laser frequency combs and stabilized etalons ensure that the instrumental stability is well characterized. However, additional sources of error include stellar noise, undetected short-period planets, and telluric contamination. To understand and ultimately mitigate error sources, the contributing terms in the error budget must be isolated to the greatest extent possible. Here, we introduce a new high cadence radial velocity program, the EXPRES 100 Earths program, which aims to identify rocky planets around bright, nearby G and K dwarfs. We also present a benchmark case: the 62-d orbit of a Saturn-mass planet orbiting the chromospherically quiet star, HD 3651. The combination of high eccentricity (0.6) and a moderately long orbital period, ensures significant dynamical clearing of any inner planets. Our Keplerian model for this planetary orbit has a residual RMS of 58 cm/s over a $\sim 6$ month time baseline. By eliminating significant contributors to the radial velocity error budget, HD 3651 serves as a standard for evaluating the long term precision of extreme precision radial velocity (EPRV) programs.

The EXtreme PREcision Spectrograph (EXPRES) is an environmentally stabilized, fiber-fed, $R=137,500$, optical spectrograph. It was recently commissioned at the 4.3-m Lowell Discovery Telescope (LDT) near Flagstaff, Arizona. The spectrograph was designed with a target radial-velocity (RV) precision of 30$\mathrm{~cm~s^{-1}}$. In addition to instrumental innovations, the EXPRES pipeline, presented here, is the first for an on-sky, optical, fiber-fed spectrograph to employ many novel techniques---including an "extended flat" fiber used for wavelength-dependent quantum efficiency characterization of the CCD, a flat-relative optimal extraction algorithm, chromatic barycentric corrections, chromatic calibration offsets, and an ultra-precise laser frequency comb for wavelength calibration. We describe the reduction, calibration, and radial-velocity analysis pipeline used for EXPRES and present an example of our current sub-meter-per-second RV measurement precision, which reaches a formal, single-measurement error of 0.3$\mathrm{~m~s^{-1}}$ for an observation with a per-pixel signal-to-noise ratio of 250. These velocities yield an orbital solution on the known exoplanet host 51 Peg that matches literature values with a residual RMS of 0.895$\mathrm{~m~s^{-1}}$.

The EXtreme PREcision Spectrograph (EXPRES) is a new Doppler spectrograph designed to reach a radial velocity measurement precision sufficient to detect Earth-like exoplanets orbiting nearby, bright stars. We report on extensive laboratory testing and on-sky observations to quantitatively assess the instrumental radial velocity measurement precision of EXPRES, with a focused discussion of individual terms in the instrument error budget. We find that EXPRES can reach a single-measurement instrument calibration precision better than 10 cm/s, not including photon noise from stellar observations. We also report on the performance of the various environmental, mechanical, and optical subsystems of EXPRES, assessing any contributions to radial velocity error. For atmospheric and telescope related effects, this includes the fast tip-tilt guiding system, atmospheric dispersion compensation, and the chromatic exposure meter. For instrument calibration, this includes the laser frequency comb (LFC), flat-field light source, CCD detector, and effects in the optical fibers. Modal noise is mitigated to a negligible level via a chaotic fiber agitator, which is especially important for wavelength calibration with the LFC. Regarding detector effects, we empirically assess the impact on radial velocity precision due to pixel-position non-uniformities (PPNU) and charge transfer inefficiency (CTI). EXPRES has begun its science survey to discover exoplanets orbiting G-dwarf and K-dwarf stars, in addition to transit spectroscopy and measurements of the Rossiter-McLaughlin effect.

When applying the barycentric correction to a precise radial velocity measurement, it is common practice to calculate its value only at the photon-weighted midpoint time of the observation instead of integrating over the entire exposure. However, since the barycentric correction does not change linearly with time, this leads to systematic errors in the derived radial velocities. The typical magnitude of this second-order effect is of order 10 cm s$^{-1}$, but it depends on several parameters, e.g. the latitude of the observatory, the position of the target on the sky, and the exposure time. We show that there are realistic observing scenarios, where the errors can amount to more than 1 ms$^{-1}$. We therefore recommend that instruments operating in this regime always record and store the exposure meter flux curve (or a similar measure) to be used as photon-weights for the barycentric correction. In existing data, if the flux curve is no longer available, we argue that second-order errors in the barycentric correction can be mitigated by adding a correction term assuming constant flux.

One source of error in high-precision radial velocity measurements of exoplanet host stars is chromatic change in Earth's atmospheric transmission during observations. Mitigation of this error requires that the photon-weighted barycentric correction be applied as a function of wavelength across the stellar spectrum. We have designed a system for chromatic photon-weighted barycentric corrections with the EXtreme PREcision Spectrograph (EXPRES) and present results from the first year of operations, based on radial velocity measurements of more than $10^3$ high-resolution stellar spectra. For observation times longer than 250 seconds, we find that if the chromatic component of the barycentric corrections is ignored, a range of radial velocity errors up to 1 m s$^{-1}$ can be incurred with cross-correlation, depending on the nightly atmospheric conditions. For this distribution of errors, the standard deviation is 8.4 cm s$^{-1}$ for G-type stars, 8.5 cm s$^{-1}$ for K-type stars, and 2.1 cm s$^{-1}$ for M-type stars. This error is reduced to well-below the instrumental and photon-noise limited floor by frequent flux sampling of the observed star with a low-resolution exposure meter spectrograph.

Atmospheric effects on stellar radial velocity measurements for exoplanet discovery and characterization have not yet been fully investigated for extreme precision levels. We carry out calculations to determine the wavelength dependence of barycentric corrections across optical wavelengths, due to the ubiquitous variations in air mass during observations. We demonstrate that radial velocity errors of at least several cm/s can be incurred if the wavelength dependence is not included in the photon-weighted barycentric corrections. A minimum of four wavelength channels across optical spectra (380-680 nm) are required to account for this effect at the 10 cm/s level, with polynomial fits of the barycentric corrections applied to cover all wavelengths. Additional channels may be required in poor observing conditions or to avoid strong telluric absorption features. Furthermore, consistent flux sampling on the order of seconds throughout the observation is necessary to ensure that accurate photon weights are obtained. Finally, we describe how a multiple-channel exposure meter will be implemented in the EXtreme PREcision Spectrograph (EXPRES).